1 Source of this tutorial

R package: simpleSingleCell
Chapter: Analyzing single-cell RNA-seq data containing UMI counts

Aaron T. L. Lun
- Cancer Research UK Cambridge Institute, Li Ka Shing Centre, Robinson Way, Cambridge CB2 0RE, United Kingdom (CRUK)
Davis J. McCarthy
- EMBL European Bioinformatics Institute, Wellcome Genome Campus, Hinxton, Cambridge CB10 1SD, United Kingdom (EMBL), St Vincent’s Institute of Medical Research, 41 Victoria Parade, Fitzroy, Victoria 3065, Australia
John C. Marioni
- CRUK, EMBL, Wellcome Trust Sanger Institute, Wellcome Genome Campus, Hinxton, Cambridge CB10 1SA, United Kingdom

2 Advantages of single-cell RNA-seq

Preserves heterogeneity of cell types in a tissue
Can provide insight into the number of different cell types in a tissue
Can distinguish previously unknown cellular subtypes
Offers opportunity for more targeted analysis of cells

3 SingleCellExperiment

S4 class available from Bioconductor for single-cell RNAseq data
To install for Bioconductor version 3.8 or later (released October 2018; works with R version 3.5.0 or later):

# run only once
if (!requireNamespace("BiocManager"))
     install.packages("BiocManager")
BiocManager::install("SingleCellExperiment")
BiocManager::install("simpleSingleCell", version = "3.8")
BiocManager::install("scater")

4 Sample single-cell dataset

Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq

Amit Zeisel, Ana B. Muñoz Manchado, Peter Lönnerberg, Gioele La Manno, Simone Codeluppi, Anna Juréus, Sueli Marques, Hermany Munguba, Liqun He, Christer Betsholtz, Charlotte Rolny, Gonçalo Castelo-Branco, Jens Hjerling-Leffler and Sten Linnarsson

Science 347:1138-1142, 2015

http://linnarssonlab.org/cortex/

5 Preparation of data

Import data from Linnarsson Lab website
- Reads are already filtered for quality and mapped
- Data include mRNA, mitochondrial genes, and ERCC spike-ins (control)
Clean up data and combine datasets
Add ENSEMBL gene names
Create sce object that contains count matrices and metadata

6 Quality control overview

##   ByLibSize ByFeature BySpike Remaining
## 1         8         3       8      2989

7 Gene-level metrics

# Gene abundance # Normalization of cell-specific biases # Modeling and removing technical noise # Distribution of expression of most highly expressed genes # Dimensionality reduction # Graph-based clustering

## my.clusters
##   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16 
## 194 104 237 633 147 238 634 160 156  91  17 173  84  57  27  37

# Evaluating graph-based clustering # Detecting subpopulation-specific marker genes

## DataFrame with 10 rows and 8 columns
##              Top               p.value                   FDR
##        <integer>             <numeric>             <numeric>
## Snap25         1 1.07760786252372e-273 2.13765071688825e-269
## Stmn3          1 8.01751747584536e-239 7.95217470841747e-235
## Mllt11         1 3.58429115569114e-202 1.42203167310888e-198
## Gad1           1 1.52719585816084e-177 3.02949842383362e-174
## Atp1a3         1 5.02531526306102e-173 8.30726490611157e-170
## Celf4          1 1.85709125842012e-162 2.16700701725171e-159
## Gad2           1 2.34227647329593e-152 2.11198810912598e-149
## Rcan2          1  1.37388316275679e-92  2.99491431863808e-90
## Synpr          1   1.0227847880535e-65  8.34937524305225e-64
## Tspyl4         2 1.47394127859716e-155  1.5388722707122e-152
##                   logFC.2          logFC.3           logFC.4
##                 <numeric>        <numeric>         <numeric>
## Snap25   1.42027121155466 3.80074765847389 0.605025633438071
## Stmn3    1.95773993790992 4.33436131857807 0.190025402712248
## Mllt11   1.65026134005567 3.08053402764227 0.485210901310954
## Gad1     3.96850838614864 4.03241470752414  4.12496737409042
## Atp1a3 0.0608442678448413 3.27602693247929 0.201568491665177
## Celf4   0.518085570255083 2.72886122621097  0.54708311844668
## Gad2     3.50737849218741 3.67858313779067  3.79238727543879
## Rcan2    2.17016391558053 1.81756905217195  2.22661702113425
## Synpr    3.04399639667943 3.11252157099165   3.0631569369004
## Tspyl4   1.24441455287679 2.78590837300274  1.36491453676626
##                  logFC.5             logFC.6
##                <numeric>           <numeric>
## Snap25 0.628791789169103  -0.468427774667559
## Stmn3   1.03511245025969 -0.0939076782690842
## Mllt11  1.17528301247514   0.418382860258258
## Gad1    3.75237913044294    4.15245102170302
## Atp1a3  1.21412773443475    1.01986241645697
## Celf4  0.850310787576184   0.318147219022062
## Gad2    3.47741412338435    3.80697721719507
## Rcan2   1.55046954978362    1.19099658493111
## Synpr    2.6477236735327    2.85526954311159
## Tspyl4  1.55432426140375     1.5044419890837

Single-cell RNA-seq using available data

26 Jun 2019

Contents